"Handshake" matching MCP servers:

Enables AI agents to list, retrieve, add, and audit credentials from an encrypted vault during an active human-unlocked session, while providing no tool to unlock the vault itself.

Serves a stateless Streamable-HTTP endpoint with an echo tool, letting users observe exactly how many HTTP requests are involved in a single MCP tool call.

Enables running a minimal MCP server over stdio that manually implements JSON-RPC, exposing an add tool so users can observe the handshake, tool listing, and tool call cycle without the official SDK.

A learning-oriented MCP server that exposes basic tools (echo, add, reverse) over stdio transport to validate MCP handshake and tool calls.

A remote MCP server that gives MCP clients authenticated access to GitHub: repos, issues, pull requests, file contents, and search, over a proper OAuth 2.1 handshake. Runs on Cloudflare Workers.

Enables multiple MCP clients to securely access various third-party MCP backends through a single HTTP gateway, with modern MCP handshake compatibility and session lifecycle management.

Enables LLM agents to interact directly with Qualcomm EDL devices over USB, including device enumeration, Sahara protocol handshake, SoC identity queries, Firehose programmer upload, command-mode execution, and device reset.

A beginner-friendly MCP-inspired JSON-RPC server built with Node.js, offering basic client-server interaction through an 'initialize' capabilities handshake and an 'echo' function.

Enables agents to search MCP servers ranked by measured handshake checks, tool lists, repository signals, and signed reports, with open data.

Enables pre-2026-07-28 MCP clients using the old initialize handshake to communicate with 2026-07-28 MCP servers by acting as a proxy, with no changes to either side.

Enables local tool calling over Model Context Protocol via stdio, providing deterministic tools such as calc.add, text.word_count, and text.summarize_naive after JSON-RPC handshake and discovery.

Wraps a legacy MCP stdio server and presents it as a modern stateless 2026-07-28 protocol server, translating the legacy handshake and server-initiated requests.

Enables remote execution of basic arithmetic operations (add and subtract) via MCP over HTTP/SSE in a distributed network, demonstrating MCP handshake, capability negotiation, and dynamic tool discovery.

A base template for instantiating TypeScript stdio MCP servers, providing a single registerTools seam, structured output, and annotation presets that force every tool to carry its safety hints. It ships a get_datetime example tool that exercises the SDK end to end — handshake, annotations, structured output, and schema rejection — plus tool-level tests, so new servers inherit a proven scaffold rather than discovering SDK unknowns themselves.

Enables AI assistants to verify single email addresses and bulk lists of up to 1,000 addresses in real time, checking RFC 5322 syntax, DNS MX records, disposable providers, role accounts, and typos, with an optional deep SMTP mailbox handshake. Bulk runs handle partial-quota situations gracefully.

Enables MCP clients such as Claude Code, Codex and Antigravity to connect to a single always-on local service instead of spawning their own copy of every stdio server. Each window's handshake is served from cached memory with no process launched, and only on the first real call does that window get its own private server process — replayed faithfully, isolated per window, and stopped when the window closes.

Enables agents to inspect and diagnose their own MCP setup by exposing read-only tools that audit configured servers, report which coding agents have configs, surface misplaced or duplicated registrations, and list curated preset packs. It also powers CLI workflows for checking handshake, tool lint, and tool-surface quality scores of any stdio server.

Launches any MCP server as a real client would, running the handshake, inspecting every tool's schema and description, and deliberately making a bad tool call to test error handling. Returns one deterministic, evidence-backed score with named issues so servers can be verified before users' agents hit them in production.

Enables AI agents such as Claude Code, Cursor, and Codex to safely drive a locally installed Xilinx Vivado through 32 curated tools spanning GUI, Tcl, and attach sessions, synthesis, implementation, timing, CDC, XDC, IP, waveforms, and device programming. It provides structured timing evidence with baseline comparison, offline VCD queries for unknown values and handshake events, environment self-checks, and Chinese-language diagnosis of timing issues and CRITICAL WARNINGs.